The primary aim of the this collaborative R01 application is to conduct a longitudinal neuroimaging study of developmental changes in the function and structure of ventral-prefrontal limbic cortical networks key to emotion processing in a unique cohort of children with preschool onset Major Depressive Disorder (PO-MDD) as they transition from school age to early adolescence. To achieve this goal, we propose to conduct structural and functional neuroimaging at 3 timepoints in healthy and PO-MDD children ascertained as part of an ongoing R01 to Joan Luby, M.D. The proposed study would capture children at the 5th and 6th (final) annual waves of data collection in this continuing R01 and would conduct an additional scan and diagnostic assessment at a subsequent planned wave. Detailed longitudinal diagnostic, psychosocial, biological and family history and parenting data are already available in the study sample. The proposed study provides an invaluable opportunity to examine the effects of depression severity and course, as well as key biological and psychosocial mediators/moderators assessed during early childhood, on the structure and function of prefrontal limbic emotion systems known to be altered in older children and adults with MDD. It will also provide the first available longitudinal data on alterations in brain maturation in childhood MDD. Preliminary neuroimaging data from the study population have been obtained as part of a small NIMH supplement and support several of the specific hypotheses. To characterize differences in regional volume and cortical integrity based on innovative high dimensional computational anatomy approaches, we will focus on the neuromorphometry of the following regions: the ventral medial prefrontal cortex, pregenual cingulate, amygdala and hippocampus. Diffusion tensor imaging (DTI) will also be employed in a complementary fashion to further characterize cortical, subcortical and white matter tract structural integrity. To characterize differences in the functional neuroanatomy of emotion processing in these same regions, we propose to use functional magnetic resonance imaging during an emotion recognition paradigm and to assess functional connectivity. Based on the design of the proposed study and the available well-studied sample with PO-MDD, we will have the unique opportunity to fill a critical gap in the literature and investigate developmental changes in brain maturation related to childhood onset MDD.